Stakeholder Values and Decision Legitimacy: How to Make Defensible Decisions

Last Updated June 6, 2026

Stakeholder Values and Decision Legitimacy examines how decisions become publicly, institutionally, and ethically defensible when affected groups disagree about values, priorities, trade-offs, evidence, and acceptable risk. In decision science, stakeholder values are not an optional communication layer added after analysis. They shape what counts as a good outcome, whose burdens matter, which trade-offs are acceptable, and whether a decision can be justified to the people affected by it.

Stakeholder Values and Decision Legitimacy connects decision analysis, stakeholder theory, multi-criteria decision analysis, procedural justice, public reason, governance, accountability, value-focused thinking, ethical trade-offs, participatory decision-making, legitimacy theory, decision records, and decision-making under deep uncertainty. Its central argument is that technically strong decisions can still fail when they ignore who is affected, whose values are represented, how trade-offs are justified, and whether the process gives stakeholders a meaningful basis for trust, contestation, and accountability.

Many decision failures are not failures of calculation. They are failures of legitimacy. A model may be technically competent, a forecast may be well calibrated, and a cost-benefit analysis may be internally consistent, yet the final decision may still be rejected because affected people believe the process ignored them, misrepresented their values, shifted burdens unfairly, concealed trade-offs, or treated public justification as an afterthought.

Decision science must therefore address more than evidence and optimization. It must also examine whose values count, how conflicts are structured, how trade-offs are explained, and how decisions remain accountable when no option satisfies everyone. This is especially important in public policy, infrastructure, healthcare, sustainability, AI governance, crisis management, organizational strategy, and long-horizon planning, where decisions distribute benefits and burdens across different groups.

Stakeholder legitimacy does not mean every stakeholder gets a veto. It does not mean all preferences are equally justified. It does not mean technical evidence should be replaced by popularity. It means that decisions affecting people should be made through processes that can explain why certain values were prioritized, why certain burdens were accepted, why alternatives were rejected, and how affected groups can challenge, understand, or review the decision.

Why Stakeholder Values Matter

Stakeholder values matter because decisions are rarely evaluated only by aggregate performance. A policy may increase total welfare while imposing concentrated harm. A hospital protocol may improve efficiency while weakening patient autonomy. An infrastructure plan may reduce long-term risk while disrupting one community. An AI system may improve throughput while creating opacity, contestability problems, or disproportionate error for affected groups.

Stakeholder values define what counts as benefit, harm, fairness, dignity, safety, autonomy, resilience, efficiency, legitimacy, and acceptable risk. Without them, decision analysis can appear neutral while silently embedding the values of modelers, sponsors, regulators, funders, executives, or dominant groups.

Stakeholder values also matter because decisions often depend on cooperation. A decision that affected people view as illegitimate may face noncompliance, resistance, litigation, reputational damage, implementation failure, political reversal, or institutional distrust. Legitimacy is therefore not merely ethical. It is operational.

Stakeholder values matter because decision quality is not only a property of the selected option. It is also a property of the justification behind the option.

Back to top ↑

What Are Stakeholder Values?

Stakeholder values are the principles, priorities, concerns, standards, rights, obligations, and lived consequences that affected groups bring to a decision. They are not identical to preferences. A stakeholder may prefer one outcome but value fairness, dignity, transparency, safety, or autonomy enough to accept a different outcome if the process is justified.

Values can be explicit or implicit. They may appear in formal rules, public comments, ethical commitments, professional standards, community histories, operational constraints, lived experience, or political claims. Some values are measurable. Others require interpretation, deliberation, and explanation.

Decision science treats stakeholder values as decision-relevant because they shape objectives, criteria, weights, thresholds, constraints, unacceptable outcomes, trade-off rules, monitoring requirements, and accountability mechanisms.

Stakeholder values turn decision analysis from a calculation of outcomes into a structured account of what outcomes mean and why they matter.

Back to top ↑

What Is Decision Legitimacy?

Decision legitimacy is the extent to which a decision is accepted as appropriate, justified, authorized, fair, and accountable within a relevant social, institutional, legal, or ethical context. A legitimate decision is not necessarily popular. It is a decision that can be defended through reasons, process, evidence, authority, and value justification.

Legitimacy has several dimensions. A decision may be procedurally legitimate because the process was fair, transparent, inclusive, and reviewable. It may be substantively legitimate because the outcome respects important values, rights, and thresholds. It may be epistemically legitimate because evidence and expertise were used responsibly. It may be institutionally legitimate because the decision-maker had authority and followed appropriate rules.

These forms can conflict. A decision may follow formal rules but ignore affected groups. It may produce a beneficial outcome through an opaque or coercive process. It may use strong evidence while embedding unfair value assumptions. Strong decision legitimacy requires attention to all of these dimensions.

Decision legitimacy is not achieved by declaring a decision evidence-based. It is achieved by showing how evidence, values, authority, and accountability were brought together responsibly.

Back to top ↑

Values, Interests, and Preferences

Stakeholder analysis often becomes weak when it treats values, interests, and preferences as the same thing. They are related, but they serve different roles in decision-making.

Preferences are what stakeholders say they want. Interests are what affects their well-being, position, resources, rights, risks, or opportunities. Values are the reasons stakeholders believe certain interests or outcomes matter. A community may prefer one infrastructure route, have an interest in avoiding displacement, and value stability, heritage, environmental protection, or fairness.

Good decision science does not merely collect preferences. It asks what values and interests those preferences express. This matters because preferences can change when stakeholders receive better information, see trade-offs more clearly, or participate in deliberation. Values may be more stable and more important for legitimacy.

Stakeholder engagement becomes stronger when it moves from “What do you want?” to “What values, burdens, rights, risks, and reasons are at stake?”

Back to top ↑

Stakeholder Mapping and Affectedness

Stakeholder mapping identifies who is affected by a decision, who has influence over it, who holds relevant knowledge, who bears risk, who benefits, who pays, who can contest the decision, and who may be absent from the process. In legitimacy-sensitive decisions, mapping should not be limited to powerful or visible stakeholders.

Affectedness is central. Some stakeholders have formal authority but little direct exposure. Others have little authority but significant exposure to consequences. A public agency, contractor, regulator, local community, future resident, patient group, ecosystem, worker group, or marginalized population may each have different kinds of stake.

Stakeholder mapping should therefore distinguish influence from burden. A group with low influence but high exposure deserves special attention because its values may otherwise disappear from the decision structure.

A stakeholder map is not a contact list. It is a model of affectedness, influence, knowledge, vulnerability, and legitimacy.

Back to top ↑

Power, Vulnerability, and Representation

Stakeholder values cannot be interpreted fairly without considering power. Some groups can easily enter decision processes through funding, expertise, lobbying, institutional position, legal authority, or media access. Others may be affected but underrepresented because of language, disability, poverty, geography, immigration status, time constraints, distrust, technical barriers, or historical exclusion.

Decision legitimacy requires more than inviting comments. It requires asking whether participation conditions allow meaningful voice. A public meeting held at the wrong time, in technical language, without translation, without compensation, or after major decisions have already been made may be participatory in form but weak in substance.

Vulnerability also matters because equal treatment can produce unequal consequences. A minor cost to one group may be a major burden to another. A decision that appears efficient at the aggregate level may be illegitimate if it repeatedly shifts risk onto groups least able to absorb harm.

Legitimacy depends not only on who is invited into the process, but on whether their participation can actually affect the decision.

Back to top ↑

Trade-Offs and Competing Objectives

Stakeholder values often conflict because decisions involve competing objectives. Efficiency may conflict with equity. Safety may conflict with autonomy. Speed may conflict with deliberation. Innovation may conflict with accountability. Cost control may conflict with resilience. Local benefits may conflict with regional needs. Present gains may conflict with future burdens.

Legitimate decisions do not eliminate trade-offs. They make trade-offs visible, justified, and reviewable. If a decision sacrifices one value to protect another, the decision record should explain why the trade-off was accepted, who bears the burden, whether alternatives were considered, and what safeguards or compensation are required.

Some trade-offs are routine. Others are ethically difficult. Some values can be balanced through weights. Others may be constraints or thresholds that should not be traded away without extraordinary justification. Decision science must therefore distinguish weighted criteria from non-negotiable limits.

A decision becomes more legitimate when its trade-offs are not hidden inside a score, model, budget line, or executive judgment.

Back to top ↑

Procedural Legitimacy

Procedural legitimacy concerns how the decision was made. It asks whether the process was fair, transparent, inclusive, reason-giving, consistent, contestable, and accountable. People may accept difficult outcomes more readily when they believe the process treated them with respect and gave their concerns meaningful consideration.

Procedural legitimacy does not require endless participation. The appropriate process depends on the stakes, urgency, reversibility, affectedness, legal context, and complexity of the decision. A crisis decision may require faster procedures than a long-term infrastructure plan. But even urgent decisions should document values, assumptions, and review mechanisms.

Strong procedural legitimacy includes clear decision rights, early engagement, accessible information, explanation of alternatives, opportunity for objection, response to stakeholder concerns, and a visible path for review or revision.

Procedural legitimacy matters because the way a decision is made communicates whose agency, knowledge, and dignity the institution recognizes.

Back to top ↑

Substantive Legitimacy

Substantive legitimacy concerns the content of the decision. A process can be participatory and transparent while still producing an outcome that violates important values, imposes unjust burdens, ignores rights, or falls below minimum acceptable thresholds.

For this reason, stakeholder legitimacy cannot be reduced to consultation. A decision must also meet substantive standards. These may include safety thresholds, legal rights, distributive fairness, environmental limits, service continuity, human dignity, professional ethics, non-discrimination, and protection of vulnerable groups.

Substantive legitimacy is especially important when a decision harms some stakeholders for the benefit of others. In such cases, the decision must explain why the burden is justified, whether less harmful alternatives were considered, whether mitigation is adequate, and whether the affected group has access to review, remedy, or compensation.

Substantive legitimacy asks whether the decision deserves acceptance, not merely whether people were asked for input.

Back to top ↑

Epistemic Legitimacy

Epistemic legitimacy concerns the quality and responsible use of knowledge. A decision may fail legitimacy if it uses poor evidence, hides uncertainty, ignores local knowledge, overstates model confidence, excludes relevant expertise, or treats contested assumptions as settled facts.

Stakeholders often challenge decisions not only because they dislike the outcome, but because they distrust the evidence behind it. They may question data sources, model assumptions, expert independence, sampling, missing variables, error rates, or whether lived experience was dismissed as anecdotal.

Strong epistemic legitimacy requires transparent evidence, uncertainty disclosure, model comparison, explanation of assumptions, inclusion of contextual knowledge, and a clear distinction between empirical findings and value judgments.

Epistemic legitimacy means stakeholders can understand not only what evidence was used, but how uncertainty, disagreement, and judgment were handled.

Back to top ↑

Public Reason and Explanation

Legitimate decisions require reasons that can be offered to those affected. This does not mean every stakeholder will agree. It means the decision-maker can explain the decision in terms of evidence, values, authority, trade-offs, constraints, and responsibilities that affected people can inspect and challenge.

Public explanation is different from messaging. Messaging tries to persuade people to accept a decision. Explanation makes the decision intelligible. It shows what alternatives were considered, what values were prioritized, what burdens were accepted, why objections did or did not change the decision, and how the decision will be monitored or revised.

In complex decisions, explanation should be layered. Technical reports may serve analysts and experts. Public summaries may serve affected communities. Decision records may serve accountability and audit. Stakeholder-specific explanations may be needed when different groups face different risks or consequences.

Explanation is not a public-relations accessory. It is part of the decision’s legitimacy architecture.

Back to top ↑

MCDA and Stakeholder Values

Multi-criteria decision analysis, or MCDA, is useful when decisions involve multiple criteria that cannot be reduced easily to a single measure. It can help structure stakeholder values by identifying criteria, scoring alternatives, weighting priorities, testing sensitivity, and comparing trade-offs.

MCDA is especially valuable when stakeholders disagree because it makes the structure of disagreement visible. One group may prioritize cost, another safety, another equity, another autonomy, another long-term resilience. MCDA can show how rankings change under different value weights.

But MCDA can also mislead. If weights are treated as objective, if stakeholders are poorly represented, if criteria omit important values, or if low scores on rights or safety are offset too easily by high scores elsewhere, MCDA can create false legitimacy. It should be used as a structured deliberation tool, not a machine for producing moral closure.

MCDA is strongest when it helps stakeholders see and debate value trade-offs, not when it hides disagreement behind a final score.

Back to top ↑

Stakeholder Values Under Deep Uncertainty

Stakeholder values become even more important under deep uncertainty. When probabilities are unreliable, models disagree, future states are contested, and consequences unfold over long time horizons, decision-makers cannot rely only on technical optimization. They must also decide which risks are acceptable, which futures should be protected against, and whose burdens matter when uncertainty remains unresolved.

Deep uncertainty often increases legitimacy demands because affected groups may disagree about scenario plausibility, model credibility, acceptable precaution, and the fairness of waiting or acting. One stakeholder may prefer immediate action to prevent harm. Another may prefer delay to gather more evidence. Another may prioritize reversibility. Another may reject any option that violates a threshold.

In these contexts, decision legitimacy depends on making uncertainty and values visible together. Stakeholders should be able to see not only what could happen, but how different value commitments change the preferred decision.

Under deep uncertainty, legitimacy is not the elimination of disagreement. It is the disciplined management of disagreement under conditions where certainty is unavailable.

Back to top ↑

Governance and Accountability

Stakeholder legitimacy requires governance. Values must be connected to decision rights, review processes, documentation, monitoring, escalation, appeal, and learning. Otherwise stakeholder engagement becomes detached from actual authority.

A decision record should preserve the stakeholder map, affectedness analysis, value criteria, trade-offs, dissent, thresholds, evidence sources, engagement process, rationale, and review triggers. This prevents hindsight distortion and allows the decision to be audited later.

Accountability also requires clarity about which stakeholder claims changed the decision and which did not. Ignoring a stakeholder claim may be justified, but it should not be invisible. A legitimate process explains how claims were evaluated and why the final decision remains defensible.

Governance turns stakeholder values from consultation language into enforceable decision accountability.

Back to top ↑

Applications Across Decision Contexts

Stakeholder values and decision legitimacy are relevant wherever decisions distribute benefits, burdens, risk, authority, or opportunity across groups. The specific values differ by domain, but the legitimacy question remains: can the decision be justified to those affected?

Across contexts, legitimacy depends on whether affected people can see how their values, risks, and objections were considered.

Back to top ↑

Limitations and Challenges

Stakeholder-centered decision-making has limitations. Engagement can be slow, expensive, incomplete, politicized, or performative. Stakeholders may disagree with each other. Some may make claims that conflict with evidence, rights, or legal obligations. Powerful groups may dominate the process. Decision-makers may use stakeholder language to create the appearance of legitimacy without sharing real influence.

There is also a risk of false consensus. A process may produce a summary of “stakeholder input” that erases dissent, complexity, or minority concerns. Legitimacy does not require consensus, but it does require honest treatment of disagreement.

Another challenge is scale. Large systems may include many stakeholders with different levels of affectedness, influence, and knowledge. Decision-makers must avoid both extremes: ignoring stakeholders because engagement is difficult, and pretending every stakeholder claim can be satisfied.

The challenge is not simply to include stakeholders. The challenge is to design decision systems where stakeholder values can genuinely shape what is evaluated, chosen, justified, and revised.

Back to top ↑

Summary Table: Stakeholder Values and Decision Legitimacy

The table below summarizes the main concepts involved in stakeholder values and decision legitimacy.

Stakeholder legitimacy is the bridge between structured analysis and defensible authority.

Back to top ↑

Back to top ↑

Mathematical Lens: Stakeholder Weights, Burdens, Legitimacy, and Trade-Offs

The mathematical lens helps clarify how stakeholder values can be represented, compared, and tested without pretending that legitimacy is reducible to a single number.

Let \(A\) be the set of decision alternatives, \(C\) the set of criteria, and \(G\) the set of stakeholder groups. Let \(x_{a,c}\) represent the performance of alternative \(a\) on criterion \(c\).

An aggregate score can combine stakeholder scores using stakeholder weights \(\alpha_g\):

But aggregate scores can hide concentrated harm. A burden function can track negative impact for each group:

A max-burden rule can be used to evaluate concentrated harm:

A threshold rule can identify whether any stakeholder group falls below an acceptable standard:

A procedural legitimacy score can include participation, transparency, explanation, contestability, and review:

A decision legitimacy index can combine substantive and procedural factors:

The mathematical lesson is that stakeholder legitimacy cannot be collapsed into one score without loss. But structured scoring can still reveal whose values dominate, whose burdens are hidden, and which alternatives fail legitimacy thresholds.

Back to top ↑

R Workflow: Stakeholder Values, Trade-Offs, and Legitimacy Scores

The R workflow below compares alternatives across stakeholder value profiles, procedural legitimacy criteria, burden measures, threshold checks, and decision legitimacy scores. It uses base R so it can run without additional package installation.

# stakeholder_values_decision_legitimacy_workflow.R
# Base R workflow for stakeholder values and decision legitimacy:
# stakeholder-specific value scores, burden analysis, legitimacy thresholds,
# procedural scores, decision legitimacy index, and decision records.

args <- commandArgs(trailingOnly = FALSE)
file_arg <- grep("^--file=", args, value = TRUE)

if (length(file_arg) > 0) {
  script_path <- normalizePath(sub("^--file=", "", file_arg[1]), mustWork = TRUE)
  article_root <- normalizePath(file.path(dirname(script_path), ".."), mustWork = TRUE)
} else {
  article_root <- getwd()
}

setwd(article_root)

tables_dir <- file.path(article_root, "outputs", "tables")
figures_dir <- file.path(article_root, "outputs", "figures")

dir.create(tables_dir, recursive = TRUE, showWarnings = FALSE)
dir.create(figures_dir, recursive = TRUE, showWarnings = FALSE)

alternatives <- data.frame(
  alternative = c(
    "Efficiency First",
    "Balanced Public Value",
    "Equity Protective",
    "Participatory Adaptive",
    "Precautionary Safeguard"
  ),
  cost_efficiency = c(0.92, 0.76, 0.58, 0.68, 0.62),
  service_quality = c(0.74, 0.82, 0.76, 0.80, 0.78),
  equity = c(0.38, 0.74, 0.91, 0.84, 0.86),
  autonomy = c(0.44, 0.72, 0.70, 0.82, 0.68),
  resilience = c(0.52, 0.78, 0.75, 0.86, 0.91),
  transparency = c(0.46, 0.76, 0.72, 0.90, 0.82),
  stringsAsFactors = FALSE
)

criteria <- c(
  "cost_efficiency",
  "service_quality",
  "equity",
  "autonomy",
  "resilience",
  "transparency"
)

stakeholder_weights <- data.frame(
  stakeholder = rep(c("Residents", "Service Users", "Workers", "Regulators", "Future Stakeholders"), each = length(criteria)),
  criterion = rep(criteria, times = 5),
  weight = c(
    0.12, 0.18, 0.28, 0.14, 0.16, 0.12,
    0.10, 0.30, 0.18, 0.20, 0.12, 0.10,
    0.12, 0.18, 0.20, 0.14, 0.16, 0.20,
    0.16, 0.18, 0.18, 0.12, 0.16, 0.20,
    0.08, 0.12, 0.20, 0.10, 0.34, 0.16
  ),
  stringsAsFactors = FALSE
)

stakeholder_importance <- data.frame(
  stakeholder = c("Residents", "Service Users", "Workers", "Regulators", "Future Stakeholders"),
  importance = c(0.24, 0.24, 0.18, 0.18, 0.16),
  minimum_threshold = c(0.66, 0.68, 0.64, 0.66, 0.62),
  stringsAsFactors = FALSE
)

burdens <- data.frame(
  alternative = rep(alternatives$alternative, each = 5),
  stakeholder = rep(stakeholder_importance$stakeholder, times = nrow(alternatives)),
  burden = c(
    0.62, 0.44, 0.58, 0.38, 0.55,
    0.32, 0.28, 0.34, 0.24, 0.30,
    0.22, 0.24, 0.30, 0.26, 0.24,
    0.18, 0.20, 0.26, 0.18, 0.22,
    0.24, 0.28, 0.22, 0.20, 0.16
  ),
  stringsAsFactors = FALSE
)

procedure <- data.frame(
  alternative = alternatives$alternative,
  voice = c(0.40, 0.72, 0.76, 0.92, 0.78),
  transparency = c(0.46, 0.76, 0.74, 0.90, 0.84),
  explanation = c(0.42, 0.78, 0.76, 0.88, 0.84),
  contestability = c(0.36, 0.70, 0.72, 0.86, 0.80),
  review = c(0.44, 0.74, 0.76, 0.90, 0.88),
  stringsAsFactors = FALSE
)

procedure_weights <- c(
  voice = 0.24,
  transparency = 0.20,
  explanation = 0.20,
  contestability = 0.18,
  review = 0.18
)

stakeholder_scores <- list()
counter <- 1

for (i in seq_len(nrow(alternatives))) {
  alt <- alternatives[i, ]

  for (stakeholder in stakeholder_importance$stakeholder) {
    weights <- stakeholder_weights[stakeholder_weights$stakeholder == stakeholder, ]
    weights <- weights[match(criteria, weights$criterion), ]

    score <- sum(as.numeric(alt[, criteria]) * weights$weight)

    stakeholder_scores[[counter]] <- data.frame(
      alternative = alt$alternative,
      stakeholder = stakeholder,
      stakeholder_score = score,
      stringsAsFactors = FALSE
    )

    counter <- counter + 1
  }
}

stakeholder_score_table <- do.call(rbind, stakeholder_scores)

stakeholder_score_table <- merge(
  stakeholder_score_table,
  stakeholder_importance,
  by = "stakeholder"
)

stakeholder_score_table$passes_threshold <- stakeholder_score_table$stakeholder_score >= stakeholder_score_table$minimum_threshold

aggregate_scores <- aggregate(
  stakeholder_score * importance ~ alternative,
  data = stakeholder_score_table,
  FUN = sum
)

names(aggregate_scores)[2] <- "aggregate_stakeholder_score"

threshold_summary <- aggregate(
  passes_threshold ~ alternative,
  data = stakeholder_score_table,
  FUN = function(x) mean(as.numeric(x))
)

names(threshold_summary)[2] <- "stakeholder_threshold_pass_rate"

min_score_summary <- aggregate(
  stakeholder_score ~ alternative,
  data = stakeholder_score_table,
  FUN = min
)

names(min_score_summary)[2] <- "minimum_stakeholder_score"

max_burden <- aggregate(
  burden ~ alternative,
  data = burdens,
  FUN = max
)

names(max_burden)[2] <- "maximum_stakeholder_burden"

avg_burden <- aggregate(
  burden ~ alternative,
  data = burdens,
  FUN = mean
)

names(avg_burden)[2] <- "average_stakeholder_burden"

procedure$procedural_score <- as.vector(
  as.matrix(procedure[, names(procedure_weights)]) %*% procedure_weights
)

results <- merge(aggregate_scores, threshold_summary, by = "alternative")
results <- merge(results, min_score_summary, by = "alternative")
results <- merge(results, max_burden, by = "alternative")
results <- merge(results, avg_burden, by = "alternative")
results <- merge(results, procedure[, c("alternative", "procedural_score")], by = "alternative")

results$decision_legitimacy_index <- (
  0.40 * results$aggregate_stakeholder_score +
    0.24 * results$procedural_score +
    0.18 * results$stakeholder_threshold_pass_rate +
    0.10 * results$minimum_stakeholder_score -
    0.08 * results$maximum_stakeholder_burden
)

results$review_flag <- ifelse(
  results$stakeholder_threshold_pass_rate < 0.80 |
    results$maximum_stakeholder_burden > 0.50 |
    results$procedural_score < 0.65,
  "review",
  "acceptable"
)

results$rank <- rank(-results$decision_legitimacy_index, ties.method = "min")
results <- results[order(results$rank), ]

write.csv(
  alternatives,
  file.path(tables_dir, "stakeholder_alternative_scores.csv"),
  row.names = FALSE
)

write.csv(
  stakeholder_weights,
  file.path(tables_dir, "stakeholder_value_weights.csv"),
  row.names = FALSE
)

write.csv(
  stakeholder_score_table,
  file.path(tables_dir, "stakeholder_scores_by_group.csv"),
  row.names = FALSE
)

write.csv(
  burdens,
  file.path(tables_dir, "stakeholder_burden_table.csv"),
  row.names = FALSE
)

write.csv(
  procedure,
  file.path(tables_dir, "procedural_legitimacy_scores.csv"),
  row.names = FALSE
)

write.csv(
  results,
  file.path(tables_dir, "decision_legitimacy_results.csv"),
  row.names = FALSE
)

png(file.path(figures_dir, "decision_legitimacy_index_by_alternative.png"), width = 1200, height = 800)
barplot(
  results$decision_legitimacy_index,
  names.arg = results$alternative,
  las = 2,
  main = "Decision Legitimacy Index by Alternative",
  ylab = "Decision legitimacy index"
)
grid()
dev.off()

png(file.path(figures_dir, "maximum_stakeholder_burden.png"), width = 1200, height = 800)
barplot(
  results$maximum_stakeholder_burden,
  names.arg = results$alternative,
  las = 2,
  main = "Maximum Stakeholder Burden by Alternative",
  ylab = "Maximum burden"
)
grid()
dev.off()

print(results)
print(stakeholder_score_table)

This workflow shows how stakeholder values can change the evaluation of alternatives. It also shows why legitimacy cannot be inferred from aggregate performance alone: threshold failures, procedural weakness, and concentrated burdens can all require review even when an alternative scores well overall.

Back to top ↑

Python Workflow: Mapping Stakeholder Burdens and Decision Legitimacy

The Python workflow below uses only the standard library. It computes stakeholder value scores, aggregate stakeholder performance, burden exposure, procedural legitimacy, threshold compliance, decision legitimacy scores, review flags, and a decision record.

# stakeholder_values_decision_legitimacy_simulation.py
# Standard-library workflow for stakeholder values and decision legitimacy:
# stakeholder-specific scores, burden analysis, procedural legitimacy,
# threshold checks, decision legitimacy index, and decision records.

from __future__ import annotations

from pathlib import Path
import csv
import json

ARTICLE_ROOT = Path(__file__).resolve().parents[1]
TABLES = ARTICLE_ROOT / "outputs" / "tables"
RECORDS = ARTICLE_ROOT / "outputs" / "decision_records"

CRITERIA = [
    "cost_efficiency",
    "service_quality",
    "equity",
    "autonomy",
    "resilience",
    "transparency",
]

ALTERNATIVES = [
    {
        "alternative": "Efficiency First",
        "cost_efficiency": 0.92,
        "service_quality": 0.74,
        "equity": 0.38,
        "autonomy": 0.44,
        "resilience": 0.52,
        "transparency": 0.46,
    },
    {
        "alternative": "Balanced Public Value",
        "cost_efficiency": 0.76,
        "service_quality": 0.82,
        "equity": 0.74,
        "autonomy": 0.72,
        "resilience": 0.78,
        "transparency": 0.76,
    },
    {
        "alternative": "Equity Protective",
        "cost_efficiency": 0.58,
        "service_quality": 0.76,
        "equity": 0.91,
        "autonomy": 0.70,
        "resilience": 0.75,
        "transparency": 0.72,
    },
    {
        "alternative": "Participatory Adaptive",
        "cost_efficiency": 0.68,
        "service_quality": 0.80,
        "equity": 0.84,
        "autonomy": 0.82,
        "resilience": 0.86,
        "transparency": 0.90,
    },
    {
        "alternative": "Precautionary Safeguard",
        "cost_efficiency": 0.62,
        "service_quality": 0.78,
        "equity": 0.86,
        "autonomy": 0.68,
        "resilience": 0.91,
        "transparency": 0.82,
    },
]

STAKEHOLDER_WEIGHTS = {
    "Residents": {
        "cost_efficiency": 0.12,
        "service_quality": 0.18,
        "equity": 0.28,
        "autonomy": 0.14,
        "resilience": 0.16,
        "transparency": 0.12,
    },
    "Service Users": {
        "cost_efficiency": 0.10,
        "service_quality": 0.30,
        "equity": 0.18,
        "autonomy": 0.20,
        "resilience": 0.12,
        "transparency": 0.10,
    },
    "Workers": {
        "cost_efficiency": 0.12,
        "service_quality": 0.18,
        "equity": 0.20,
        "autonomy": 0.14,
        "resilience": 0.16,
        "transparency": 0.20,
    },
    "Regulators": {
        "cost_efficiency": 0.16,
        "service_quality": 0.18,
        "equity": 0.18,
        "autonomy": 0.12,
        "resilience": 0.16,
        "transparency": 0.20,
    },
    "Future Stakeholders": {
        "cost_efficiency": 0.08,
        "service_quality": 0.12,
        "equity": 0.20,
        "autonomy": 0.10,
        "resilience": 0.34,
        "transparency": 0.16,
    },
}

STAKEHOLDER_IMPORTANCE = {
    "Residents": {"importance": 0.24, "minimum_threshold": 0.66},
    "Service Users": {"importance": 0.24, "minimum_threshold": 0.68},
    "Workers": {"importance": 0.18, "minimum_threshold": 0.64},
    "Regulators": {"importance": 0.18, "minimum_threshold": 0.66},
    "Future Stakeholders": {"importance": 0.16, "minimum_threshold": 0.62},
}

BURDENS = {
    "Efficiency First": {
        "Residents": 0.62,
        "Service Users": 0.44,
        "Workers": 0.58,
        "Regulators": 0.38,
        "Future Stakeholders": 0.55,
    },
    "Balanced Public Value": {
        "Residents": 0.32,
        "Service Users": 0.28,
        "Workers": 0.34,
        "Regulators": 0.24,
        "Future Stakeholders": 0.30,
    },
    "Equity Protective": {
        "Residents": 0.22,
        "Service Users": 0.24,
        "Workers": 0.30,
        "Regulators": 0.26,
        "Future Stakeholders": 0.24,
    },
    "Participatory Adaptive": {
        "Residents": 0.18,
        "Service Users": 0.20,
        "Workers": 0.26,
        "Regulators": 0.18,
        "Future Stakeholders": 0.22,
    },
    "Precautionary Safeguard": {
        "Residents": 0.24,
        "Service Users": 0.28,
        "Workers": 0.22,
        "Regulators": 0.20,
        "Future Stakeholders": 0.16,
    },
}

PROCEDURE = {
    "Efficiency First": {
        "voice": 0.40,
        "transparency": 0.46,
        "explanation": 0.42,
        "contestability": 0.36,
        "review": 0.44,
    },
    "Balanced Public Value": {
        "voice": 0.72,
        "transparency": 0.76,
        "explanation": 0.78,
        "contestability": 0.70,
        "review": 0.74,
    },
    "Equity Protective": {
        "voice": 0.76,
        "transparency": 0.74,
        "explanation": 0.76,
        "contestability": 0.72,
        "review": 0.76,
    },
    "Participatory Adaptive": {
        "voice": 0.92,
        "transparency": 0.90,
        "explanation": 0.88,
        "contestability": 0.86,
        "review": 0.90,
    },
    "Precautionary Safeguard": {
        "voice": 0.78,
        "transparency": 0.84,
        "explanation": 0.84,
        "contestability": 0.80,
        "review": 0.88,
    },
}

PROCEDURE_WEIGHTS = {
    "voice": 0.24,
    "transparency": 0.20,
    "explanation": 0.20,
    "contestability": 0.18,
    "review": 0.18,
}


def validate_weights() -> None:
    for stakeholder, weights in STAKEHOLDER_WEIGHTS.items():
        total = sum(weights.values())
        if abs(total - 1.0) > 1e-9:
            raise ValueError(f"Stakeholder weights must sum to 1 for {stakeholder}. Got {total}.")
    if abs(sum(PROCEDURE_WEIGHTS.values()) - 1.0) > 1e-9:
        raise ValueError("Procedure weights must sum to 1.")


def stakeholder_score(alternative: dict[str, object], stakeholder: str) -> float:
    weights = STAKEHOLDER_WEIGHTS[stakeholder]
    return sum(float(alternative[criterion]) * weights[criterion] for criterion in CRITERIA)


def procedural_score(alternative_name: str) -> float:
    values = PROCEDURE[alternative_name]
    return sum(values[key] * PROCEDURE_WEIGHTS[key] for key in PROCEDURE_WEIGHTS)


def compute_scores() -> tuple[list[dict[str, object]], list[dict[str, object]], list[dict[str, object]]]:
    stakeholder_rows: list[dict[str, object]] = []
    burden_rows: list[dict[str, object]] = []
    result_rows: list[dict[str, object]] = []

    for alternative in ALTERNATIVES:
        alternative_name = str(alternative["alternative"])
        aggregate_score = 0.0
        scores_for_thresholds = []

        for stakeholder, info in STAKEHOLDER_IMPORTANCE.items():
            score = stakeholder_score(alternative, stakeholder)
            threshold = info["minimum_threshold"]
            passes = score >= threshold

            aggregate_score += score * info["importance"]
            scores_for_thresholds.append(score)

            stakeholder_rows.append({
                "alternative": alternative_name,
                "stakeholder": stakeholder,
                "stakeholder_score": round(score, 6),
                "importance": info["importance"],
                "minimum_threshold": threshold,
                "passes_threshold": passes,
            })

        for stakeholder, burden in BURDENS[alternative_name].items():
            burden_rows.append({
                "alternative": alternative_name,
                "stakeholder": stakeholder,
                "burden": burden,
            })

        max_burden = max(BURDENS[alternative_name].values())
        average_burden = sum(BURDENS[alternative_name].values()) / len(BURDENS[alternative_name])
        pass_rate = sum(1 for row in stakeholder_rows if row["alternative"] == alternative_name and row["passes_threshold"]) / len(STAKEHOLDER_IMPORTANCE)
        min_score = min(scores_for_thresholds)
        proc_score = procedural_score(alternative_name)

        decision_legitimacy_index = (
            0.40 * aggregate_score
            + 0.24 * proc_score
            + 0.18 * pass_rate
            + 0.10 * min_score
            - 0.08 * max_burden
        )

        review = pass_rate < 0.80 or max_burden > 0.50 or proc_score < 0.65

        result_rows.append({
            "alternative": alternative_name,
            "aggregate_stakeholder_score": round(aggregate_score, 6),
            "stakeholder_threshold_pass_rate": round(pass_rate, 6),
            "minimum_stakeholder_score": round(min_score, 6),
            "maximum_stakeholder_burden": round(max_burden, 6),
            "average_stakeholder_burden": round(average_burden, 6),
            "procedural_score": round(proc_score, 6),
            "decision_legitimacy_index": round(decision_legitimacy_index, 6),
            "review_flag": "review" if review else "acceptable",
        })

    ranked = sorted(result_rows, key=lambda row: float(row["decision_legitimacy_index"]), reverse=True)
    for rank, row in enumerate(ranked, start=1):
        row["rank"] = rank

    return stakeholder_rows, burden_rows, ranked


def write_csv(path: Path, rows: list[dict[str, object]]) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    if not rows:
        raise ValueError(f"No rows to write: {path}")
    with path.open("w", encoding="utf-8", newline="") as handle:
        writer = csv.DictWriter(handle, fieldnames=list(rows[0].keys()))
        writer.writeheader()
        writer.writerows(rows)


def write_json(path: Path, payload: dict[str, object]) -> None:
    path.parent.mkdir(parents=True, exist_ok=True)
    path.write_text(json.dumps(payload, indent=2), encoding="utf-8")


def main() -> None:
    validate_weights()

    stakeholder_rows, burden_rows, results = compute_scores()

    procedure_rows = [
        {"alternative": alternative, **scores, "procedural_score": round(procedural_score(alternative), 6)}
        for alternative, scores in PROCEDURE.items()
    ]

    weight_rows = [
        {"stakeholder": stakeholder, "criterion": criterion, "weight": weight}
        for stakeholder, weights in STAKEHOLDER_WEIGHTS.items()
        for criterion, weight in weights.items()
    ]

    write_csv(TABLES / "stakeholder_alternative_scores.csv", ALTERNATIVES)
    write_csv(TABLES / "stakeholder_value_weights.csv", weight_rows)
    write_csv(TABLES / "stakeholder_scores_by_group.csv", stakeholder_rows)
    write_csv(TABLES / "stakeholder_burden_table.csv", burden_rows)
    write_csv(TABLES / "procedural_legitimacy_scores.csv", procedure_rows)
    write_csv(TABLES / "decision_legitimacy_results.csv", results)

    write_json(
        RECORDS / "stakeholder_values_decision_record.json",
        {
            "article": "Stakeholder Values and Decision Legitimacy",
            "decision_context": "Comparing alternatives across stakeholder value profiles, burdens, procedural legitimacy, thresholds, and decision legitimacy scores.",
            "criteria": CRITERIA,
            "stakeholder_importance": STAKEHOLDER_IMPORTANCE,
            "procedure_weights": PROCEDURE_WEIGHTS,
            "ranked_results": results,
            "selected_alternative": results[0]["alternative"],
            "modeling_principles": [
                "Stakeholder values should shape criteria, thresholds, and trade-off review.",
                "Aggregate scores can hide concentrated stakeholder burdens.",
                "Procedural legitimacy requires voice, transparency, explanation, contestability, and review.",
                "Threshold checks prevent low stakeholder performance from being averaged away.",
                "Decision records should preserve stakeholder claims, dissent, burdens, rationale, and review triggers."
            ],
        },
    )

    print("Stakeholder values and decision legitimacy workflow complete.")
    print(TABLES / "decision_legitimacy_results.csv")
    print(RECORDS / "stakeholder_values_decision_record.json")


if __name__ == "__main__":
    main()

This workflow shows how stakeholder values can be translated into explicit criteria, weights, thresholds, burdens, and procedural measures without pretending that legitimacy is purely quantitative. The outputs support review, explanation, and challenge.

Back to top ↑

GitHub Repository

The companion repository for this article supports reproducible exploration of stakeholder values, decision legitimacy, multi-criteria evaluation, stakeholder scoring, burden analysis, procedural legitimacy, threshold checks, public justification, dissent records, and decision-record documentation.

articles/stakeholder-values-and-decision-legitimacy/
├── python/
│   ├── stakeholder_values_decision_legitimacy_simulation.py
│   ├── stakeholder_mapping.py
│   ├── value_weight_model.py
│   ├── burden_analysis.py
│   ├── procedural_legitimacy.py
│   ├── threshold_checks.py
│   ├── legitimacy_index.py
│   ├── decision_record_exporter.py
│   └── run_all_stakeholder_legitimacy_workflows.py
├── r/
│   ├── stakeholder_values_decision_legitimacy_workflow.R
│   ├── stakeholder_scores.R
│   ├── burden_tables.R
│   ├── procedural_legitimacy_tables.R
│   ├── threshold_review_tables.R
│   ├── legitimacy_summary.R
│   └── run_all_stakeholder_legitimacy_workflows.R
├── julia/
│   ├── high_performance_legitimacy_scan.jl
│   ├── stakeholder_score_model.jl
│   └── burden_threshold_model.jl
├── sql/
│   ├── schema_stakeholder_values_decision_legitimacy.sql
│   ├── stakeholders.sql
│   ├── alternatives.sql
│   ├── criteria.sql
│   ├── value_weights.sql
│   ├── burdens.sql
│   ├── procedure_scores.sql
│   ├── decision_records.sql
│   └── sample_queries.sql
├── rust/
│   └── stakeholder_legitimacy_cli.rs
├── go/
│   └── stakeholder_legitimacy_runner.go
├── cpp/
│   ├── stakeholder_score_core.cpp
│   └── legitimacy_index_core.cpp
├── fortran/
│   └── numerical_stakeholder_legitimacy_model.f90
├── c/
│   └── stakeholder_legitimacy_core.c
├── docs/
│   ├── article_notes.md
│   ├── modeling_principles.md
│   ├── stakeholder_mapping.md
│   ├── values_interests_preferences.md
│   ├── procedural_legitimacy.md
│   ├── substantive_legitimacy.md
│   ├── epistemic_legitimacy.md
│   ├── tradeoffs_and_dissent.md
│   ├── decision_records.md
│   ├── responsible_use.md
│   └── assumptions_and_limitations.md
├── data/
│   ├── synthetic_stakeholders.csv
│   ├── synthetic_alternatives.csv
│   ├── synthetic_criteria.csv
│   ├── synthetic_value_weights.csv
│   ├── synthetic_burdens.csv
│   ├── synthetic_procedure_scores.csv
│   └── synthetic_decision_records.csv
├── outputs/
│   ├── README.md
│   ├── figures/
│   ├── tables/
│   └── decision_records/
└── notebooks/
    ├── python_stakeholder_values_decision_legitimacy_walkthrough.ipynb
    └── r_stakeholder_values_decision_legitimacy_placeholder.ipynb

This repository structure reflects the article’s central argument: stakeholder legitimacy becomes actionable when values, affectedness, burdens, process quality, thresholds, dissent, and decision rationale are explicit enough to inspect, rerun, and challenge.

Back to top ↑

Back to top ↑

Common Pitfalls

Stakeholder-centered decision-making can become weak when engagement is disconnected from authority, values are reduced to preferences, and legitimacy is treated as messaging after the decision. The goal is not to make every stakeholder happy. The goal is to make the decision process and rationale defensible.

The most common mistake is treating stakeholder legitimacy as acceptance management rather than decision accountability.

Back to top ↑

Why Stakeholder Values and Decision Legitimacy Matter

Stakeholder Values and Decision Legitimacy matter because decisions are not legitimate simply because they are modeled, optimized, documented, or made by authorized institutions. Decisions become defensible when affected groups can understand how their values, risks, burdens, knowledge, and objections were considered.

Stakeholder values shape what counts as success. Legitimacy shapes whether a decision can be justified. Procedural legitimacy asks whether the process was fair and contestable. Substantive legitimacy asks whether the outcome respects values, rights, and thresholds. Epistemic legitimacy asks whether evidence was used responsibly. Governance connects all three to authority, review, and accountability.

The goal is not consensus at any cost. Some decisions remain contested because values genuinely conflict. The goal is accountable judgment: a decision process that makes affectedness visible, trade-offs explicit, evidence honest, burdens reviewable, dissent preserved, and authority answerable. In decision science, legitimacy is not separate from decision quality. It is one of its central conditions.

Back to top ↑

Related Articles

• Decision Science
• What Is Decision Science?
• Decision Quality and the Architecture of Judgment
• Decision Records and Accountable Judgment
• Trade-Offs, Values, and Competing Objectives
• Multi-Criteria Decision Analysis
• Decision-Making Under Deep Uncertainty
• Regret Analysis and Minimax Decision Rules
• Value of Information and When to Wait
• Decision-Making in Complex Systems
• Ethics of Decision Science
• Decision Science and Democratic Public Reasoning

Back to top ↑

Further Reading

• Freeman, R.E. (2010) Strategic Management: A Stakeholder Approach. Cambridge: Cambridge University Press. Available at: Cambridge University Press.
• Suchman, M.C. (1995) “Managing Legitimacy: Strategic and Institutional Approaches,” Academy of Management Review, 20(3), pp. 571–610. Available at: Academy of Management.
• Tyler, T.R. (2006) Why People Obey the Law. Princeton: Princeton University Press. Available at: JSTOR.
• Keeney, R.L. (1992) Value-Focused Thinking: A Path to Creative Decisionmaking. Cambridge, MA: Harvard University Press. Available at: Harvard University Press.
• OECD (2022) OECD Guidelines for Citizen Participation Processes. Available at: OECD.
• Government Analysis Function (2024) An Introductory Guide to Multi-Criteria Decision Analysis. Available at: UK Government Analysis Function.
• Esmail, B.A. and Geneletti, D. (2018) “Multi-criteria decision analysis for nature conservation: A review of 20 years of applications,” Methods in Ecology and Evolution. Available at: Wiley.
• Lempert, R.J., Popper, S.W. and Bankes, S.C. (2003) Shaping the Next One Hundred Years: New Methods for Quantitative, Long-Term Policy Analysis. Santa Monica, CA: RAND Corporation. Available at: RAND.

Back to top ↑

References

• Freeman, R.E. (2010) Strategic Management: A Stakeholder Approach. Cambridge: Cambridge University Press. Available at: Cambridge University Press.
• Freeman, R.E. and McVea, J. (2001) “A Stakeholder Approach to Strategic Management.” Available at: SSRN.
• Government Analysis Function (2024) An Introductory Guide to Multi-Criteria Decision Analysis. Available at: UK Government Analysis Function.
• Keeney, R.L. (1992) Value-Focused Thinking: A Path to Creative Decisionmaking. Cambridge, MA: Harvard University Press. Available at: Harvard University Press.
• Lempert, R.J., Popper, S.W. and Bankes, S.C. (2003) Shaping the Next One Hundred Years: New Methods for Quantitative, Long-Term Policy Analysis. Santa Monica, CA: RAND Corporation. Available at: RAND.
• OECD (2021) Evidence-based policy making and stakeholder engagement. Available at: OECD.
• OECD (2022) OECD Guidelines for Citizen Participation Processes. Available at: OECD.
• Suchman, M.C. (1995) “Managing Legitimacy: Strategic and Institutional Approaches,” Academy of Management Review, 20(3), pp. 571–610. Available at: Academy of Management.
• Tyler, T.R. (2006) Why People Obey the Law. Princeton: Princeton University Press. Available at: JSTOR.

Back to top ↑